'ESTON' 236 Signal Injector

[David G4EBT]

In another thread in the 'Items Wanted' section, there was mention at post #11 of the 'ESTON' signal injector:

https://www.vintage-radio.net/forum/...d.php?t=142793

I've had one of these injectors for many years - probably a cheap rally buy, and it works well, so I thought some notes might be of interest. I won't go into signal injecting or tracing techniques - it's been covered many times on the forum, particularly in reference to the excellent Vellman K7000 kit. Suffice to say that as far as I'm concered, a signal injector/tracer is second on the list of fault finding test gear after a multimeter.

My 'ESTON' t came with typewritten instructions on a strip of paper 3” x 8”which appears to have been cut from a page, and I’d guess it dates from the early 1960s. It has a low component count – 2 capacitors (plus an isolation cap), four resistors and two transistors, which I think will probably be PNP germanium types. The brand is ‘Impex’. They’re both marked ‘S30T’ and have a green spot on top. The circuit is built on a small plain drilled Paxolin board, (not a PCB) and is a simple multivibrator circuit, the frequency of oscillation determined by the choice of resistors and capacitors. The resistor values are as follows:

Marked
Value:
680R (625R) Measured Values.
680R (625R)
4k7 (4k4)
4k7 (2k2)

Unusually for carbon composition resistors, they've drifted low over the years, rather than high.

The two capacitors aren’t marked, but one measures 8nF - the other 20nF. They will originally have been the same value, (as with the two pairs of resistors) and I think it most likely that both 20nF. As with the resistors, the value is unimportant as it only has a bearing on the frequency of oscillation. The waveform as observed isn’t quite a square wave, but that’s unimportant and I guess that the lack of symmetry is due to the difference between the actual values of the 4K7 resistors and the two capacitors. There is a 500V .01uF isolation cap from the circuit board to the probe.

As can be seen from the picture of the underside of the circuit board, it’s simply cross-wired in ‘perf-board’ style, very much along the lines of a home-brew project, but nicely conceived and in a neat purpose made screen printed case.

Signal injectors widely differ in their frequency from as low as 400 Hz to as high as 2.5 kHz. This one measures 1.955 kHz.

The typewritten instructions that came with it (now very tatty) red thus:

THE ESTON SIGNAL INJECTOR

This signal injector is a two transistor square wave generator. The battery is a No 8 and as the drain is only 5mA it should last for many months. Signal injection is a technique used by most service engineers to trace faults in television and radio sets and all that is necessary is to touch the tip of the probe to the grid of the valve or base of the transistor and work back from the output stage towards the aerial. You just press the button on the injector to inject a signal into the circuit under test and if the signal is not heard in the speaker, this indicates a faulty stage.

(Verbatim copy of original instructions).

I've drawn what I think will most likely be the circuit - a fairly standard multivibrator, but I've not traced the wiring of the little circuit board, so the actual circuit may differ.

It used a 'Number 8' 3 Volt torch battery - long since obsolete so I used a 3V CR132 lithium battery in an adaptor I made, which I'll cover in another post in this thread.

There has been speculation about the Impex brand of transistors in a forum thread below, but no conclusive evidence as to who owned the brand and when:

https://www.vintage-radio.net/forum/...d.php?t=111971

The only other references I can find show that Impex brand transistor radios were made, such as these examples:

https://www.radiomuseum.org/r/unknow...ransistor.html

https://www.ebay.co.uk/sch/i.html?_nkw=impex+transistor

Some pics attached - more to follow.

[Terry_VK5TM]

Don't want to point out the obvious, but the two "4k7" resistors are actually colour coded 5k6.

Amazing what was (and still can be) accomplished with such simple test equipment.

[David G4EBT]

Attached are pictures of the waveform and frequency of the injector. As noted in the fist post, due to variances in some of the resistor and cap actual values, the waveform isn't quite square, but it's unimportant what shape it is. It's 3V peak to peak, so certainly makes its presence known in a working radio from the front end to the speaker. (As well as their fundamental frequency, injectors of this type generate harmonics up to goodness knows what frequency, so are useful for signal injecting in the RF/IF stages as well as AF).

I've attached some pics of the replacement I made for the now obsolete Number 8 3V torch battery. I used a CR132A 3 Volt lithium battery (typically £1.50 - £2.00), which fits snugly in a piece of 20mm OD plastic conduit tube (16.75mm I.D.). The CR132A is 33mm long, and the No 8 battery is approximately 73mm overall (2.875"). I used a piece of dowel as a spacer, with a cheese head 2BA brass bolt through the centre to make up the length. The attached self explanatory pictures illustrate this.

The last pic shows the waveform of the Vellman K7000, which is a sine wave, the spec of which is 'approx' 1kHz, but is actually 1.25kHz, which is of no consequence. As with the 'ESTON' it's approximately 3V peak to peak, but the K7000 amplitude is adjustable.

Hope that's of interest.

[ms660]

One of the first weapons on the list is understanding polarity!...:

You might have time to edit.

Lawrence.

[David G4EBT]

Originally Posted by Terry_VK5TM

Don't want to point out the obvious, but the two "4k7" resistors are actually colour coded 5k6.

Amazing what was (and still can be) accomplished with such simple test equipment.

Thanks for spotting that Terry - my mistake - an 'intellectual interlude' (AKA senior moment')!

[David G4EBT]

Originally Posted by ms660

One of the first weapons on the list is understanding polarity!...

You might have time to edit.

Lawrence.

Thanks for spotting that Lawrence - drew two NPN instead of PNP transistors - asleep at the wheel again!

Too late to bury the mistake - nowhere to hide, so a corrected circuit attached.

That's two mistake in one thread (so far!) - don't tell my wife - she might put me in a home!

Incidentally, I've just noted the date on the CR132 battery. Don't know if that's the 'best before date' or date of manufacture.
Either way, the date is July 2009, so it's over eight years, and is still good.

[orbanp1]

I do not want to be picky, but that corrected circuit is not going to run either!

Here, in Figure 1, is a bjt astabil multivibrator:
https://en.wikipedia.org/wiki/Multiv..._multivibrator

Peter

[David G4EBT]

But that circuit of for NPN transistors - not PNP Peter.

Maybe Lawrence - who spotted the error in my original circuit - could confirm (hopefully!), that the circuit in my post #6 is correct?

(Originally I'd drawn NPN transistors in my first sketch - not PNP).

Not having a good day so far!

[ms660]

It's not correct....Polarity.....Bias needs to be the same polarity as Vcc.....Load needs to be in the collector circuit.

Here's a schematic for a PNP one, Vcc is -ve:

www.tpub.com/neets/book9/36a.htm

Lawrence.

[David G4EBT]

Thanks Lawrence,

Oh dear - the amended circuit below should be correct and just to make sure that it is, I've bodged it together in true 'bodgertronics' fashion, so not only have technical standards fallen today here at Bleak House - practical standards have nose-dived too. Having satisfied myself that it works, it's been consigned to the bin. (First time in a long time that I've made anything that hasn't been on a purpose made PCB, housed in a bespoke comb-jointed mahogany box!). The transistors were NOS germanium PNP AC166s, which must have been at least 50 years old. One had an HFE of 130 - the other, 134, so quite closely matched. As with the original 'ESTON' the waveform isn't a square wave, and I had to increase the value of the two caps to 100nF to get the frequency down below 2kHz. As can be seen, it's 1.943 kHz.

Normal service will be resumed as soon as possible.

[AC/HL]

Looking at the photo in post 1, those capacitors look suspiciously like early low voltage electrolytics, possibly only 1uf?

[emeritus]

Thanks for the information: the instructions for mine must have vanished long ago in my parents' house moves. It gives me a better chance of getting mine going again.

[Terry_VK5TM]

A handy table for component values v frequency of a transistor multivibrator, about half way down page:

http://www.electronics-tutorials.ws/...s/astable.html

'intellectual interlude' - think I might have to borrow that, having those sort of moments with a tape measure putting up wall panelling at the local Mens Shed.

[Boulevardier]

David - I think "intellectual interludes" are unavoidable ( and perhaps a "good thing") at our age. I have so many different things to get done that I often don't know where to turn first when I get up in the morning. Currently experimenting with measuring heat sinks' thermal resistance, trying to understand Neoliberalism, trying to sort out apostrophisation, hyphenation, and other grammar in a professional medical publication, trying to source hex-head M4 bolts (not easy), trying to "understand" two cats and two grandchildren, trying to regain mental composure after reading the Guardian - it's just endless! It's not that older minds are less acute - they just have so much more stuff to be acute about - so it gets spread out a bit more thinly...

Mike

[David G4EBT]

Originally Posted by AC/HL

Looking at the photo in post 1, those capacitors look suspiciously like early low voltage electrolytics, possibly only 1uf?

Yes, I'd agree with that Bill. There are no polarity markings, but one end is 'crimped' to seemingly differentiate it from the other. I've tried a couple of 1uF caps out of curiosity, (with the same resistors) and the frequency of oscillation is only 140Hz, so they must be a good bit lower in capacity than that. As I said earlier, the originals actually measure 8nF and 20nF respectively, so doubtless they'll have changed over time. Remarkable that this little injector still performs excellently despite it being maybe 50 years old or more. The cardboard box is marked '15 shillings'. If that was circa 1960, it equates to £16.00 today when adjusted for inflation.

(Incidentally, the scope traces I posted are 'upside down', in that the delay in the rise time represents the caps charging up on each cycle).

Thanks to everyone who has posted to the thread.